I’ve been teaching science concepts for years, and one pattern shows up again and again: students can memorize formulas, pass tests, and recite definitions, but when you ask them to question the underlying assumptions, they freeze. Not because they’re not smart enough. Because they were never taught to look for assumptions in the first place.
Science education presents itself as the domain of facts, evidence, and certainty. But underneath every scientific claim is a foundation of assumptions — some stated, many unstated — and those assumptions shape what we measure, how we interpret data, and what conclusions we draw. If you don’t know the assumptions are there, you can’t evaluate whether they’re justified.
Take something as fundamental as measuring distance. When a textbook tells you to calculate the distance to an object, it assumes light travels in a straight line between your eye and that object. In most classroom situations, that’s close enough to true that it doesn’t matter. But that assumption breaks down near the ground on a hot day, over long distances through the atmosphere, or near massive gravitational fields. The formula stays the same. The assumption underneath it? That changes depending on context.
Students aren’t taught to ask: “Under what conditions does this formula actually work?” They’re taught to plug in numbers and trust the output. The distinction matters enormously when you move from idealized problems to real-world observation.
Here’s another example that drives the point home. When you learn about the scientific method in school, you learn a clean, linear process: observation, hypothesis, experiment, conclusion. It’s presented as a neutral framework that leads reliably to truth if followed correctly.
What you don’t learn is that observation itself is theory-laden. You can’t observe anything without some framework for interpreting what you’re looking at. The tools you use, the measurements you take, the data you record — all of those choices are shaped by prior assumptions about what’s relevant, what’s measurable, and what counts as evidence.
This isn’t a flaw. It’s just how inquiry works. But it becomes a problem when students believe they’re engaging in assumption-free empiricism. They’re not. Nobody is. The question is whether the assumptions are made explicit and examined, or whether they stay invisible and unquestioned.
The role of models is particularly revealing here. Science education relies heavily on models — simplified representations of complex systems. The solar system model with planets as little balls orbiting a sun. The Bohr model of the atom with electrons in fixed orbits. The billiard-ball model of molecular collisions.
These models are useful. They help students build intuition and make predictions. But they’re not reality. They’re approximations that work under specific conditions and break down outside those conditions. The problem arises when students internalize the model as the actual thing, rather than as a tool with limitations.
I’ve watched students struggle with this transition. They learn the Bohr model in introductory chemistry, then have to unlearn parts of it when they encounter quantum mechanics. They learn Newtonian physics, then find out it’s an approximation that fails at high speeds or strong gravity. Each time, there’s a moment of disorientation. “Wait, so what I learned before was wrong?”
Not wrong. Incomplete. Context-dependent. True enough for the situations it was designed to handle.
The fault isn’t in using simplified models. The fault is in teaching those models without clearly stating their scope and limitations. When students think they’re learning universal truths instead of useful approximations, they’re being set up for confusion down the line.
Then there’s the issue of instrumentation. Science depends on instruments — thermometers, scales, telescopes, sensors — and every instrument introduces assumptions about what it’s measuring and how.
A thermometer assumes thermal equilibrium between the measuring device and the environment. A scale assumes uniform gravitational field strength. A telescope assumes a stable optical path between the object and the observer. In most cases, these assumptions are reasonable. But “reasonable” and “always true” are not the same thing.
Students are rarely taught to ask: “What is this instrument assuming? Under what conditions might those assumptions fail?” They’re taught to read the number and write it down. The underlying epistemology — the question of how we know what we know — gets skipped entirely.
This matters because assumptions aren’t just philosophical technicalities. They have practical consequences for how we interpret results.
Consider the famous Michelson-Morley experiment from 1887, designed to detect the “luminiferous aether” that light was supposedly traveling through. The experiment found no evidence of the aether. That null result was interpreted, eventually, as evidence that the aether didn’t exist. But that interpretation required rejecting a deeply held assumption about how light propagates.
The data itself was straightforward. The interpretation — what it meant — depended entirely on which assumptions you were willing to question. For years, physicists tried to save the aether hypothesis by adding additional assumptions. It took Einstein’s willingness to reject the whole framework for relativity to emerge.
Students learn about the Michelson-Morley experiment as a historical curiosity, but they don’t learn the deeper lesson: sometimes the framework itself is the problem, not the data.
So what does this mean for how science should be taught?
Start by making assumptions explicit. When you present a formula, state the conditions under which it applies. When you use a model, clarify what it simplifies and what it leaves out. When you describe an experiment, identify the underlying assumptions about measurement, observation, and interpretation.
Encourage students to ask uncomfortable questions. “How do we know that?” “What would happen if this assumption were wrong?” “Is there another way to interpret this data?” These aren’t signs of confusion. They’re signs of real scientific thinking.
And perhaps most importantly, teach the history. Show students that science is a process of refining assumptions, not a collection of eternal truths. The geocentric model worked for navigation until better observations required a heliocentric model. Newtonian mechanics worked for centuries until relativity and quantum mechanics revealed its limits. Every major advance in science involved questioning an assumption that had previously seemed unquestionable.
The goal isn’t to make students distrust science. The goal is to help them understand how science actually works — as a method of inquiry that depends on assumptions, tests them, and revises them when evidence demands it.
That’s a much richer, much more honest picture of scientific practice than the sanitized version typically presented in textbooks. And it’s the version students deserve to learn.
